RefereeBench - Entity Perception: leaderboard

Metric: Accuracy (%) on entity-perception questions (Q6), averaged across 11 sports; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 11 models tracked.

Top models

#ModelScore
1Gemini 3 Flash79.2
2Gemini 3 Pro78.6
3Seed 1.876.8
4GPT-572.6
5GPT-4o65.3
6Claude Sonnet 4.559.1
7Claude Haiku 4.555.4

Interactive version: theaggregate.ai/benchmark?slug=refereebench-entity-perception · How It Works · Data refreshed daily, snapshot 2026-10-07.